Self-Organized Criticality and Cognitive Control Reasoned by Effort Minimization

Abstract

We put forward a novel model for self-organized criticality in the dynamics of systems controlled by human actions. The model is based on two premises. First, without human control, the system in issue undergoes supercritical instability. Second, the subject’s actions are aimed at preventing the occurrence of critical fluctuations when the risk of control failure becomes essential rather than keeping the system in the stability region. The latter premise is reasoned as follows: (i) keeping the system rather far from the instability boundary is not justified from the standpoint of effort minimization, and (ii) keeping it in the immediate proximity to the instability onset also requires considerable effort because of the bounded capacity of human cognition. The concept of dynamical traps is used in the mathematical description of this type of subject’s behavior. Numerical simulation demonstrates that the proposed model does predict the emergence of fluctuations with the power-law distribution. In conclusion, we discuss that the self-organized criticality of social systems is possible due to the basic features of the human mind.